Dynamic simulation of gas turbines via feature similarity-based transfer learning

Since gas turbine plays a key role in electricity power generating, the requirements on the safety and reliability of this classical thermal system are becoming gradually strict. With a large amount of renewable energy being integrated into the power grid, the request of deep peak load regulation for satisfying the varying demand of users and maintaining the stability of the whole power grid leads to more unstable working conditions of gas turbines. The startup, shutdown, and load fluctuation are dominating the operating condition of gas turbines. Hence simulating and analyzing the dynamic behavior of the engines under such instable working conditions are important in improving their design, operation, and maintenance. However, conventional dynamic simulation methods based on the physic differential equations is unable to tackle the uncertainty and noise when faced with variant real-world operations. Although data-driven simulating methods, to some extent, can mitigate the problem, it is impossible to perform simulations with insufficient data. To tackle the issue, a novel transfer learning framework is proposed to transfer the knowledge from the physics equation domain to the real-world application domain to compensate for the lack of data. A strong dynamic operating data set with steep slope signals is created based on physics equations and then a feature similarity-based learning model with an encoder and a decoder is built and trained to achieve feature adaptive knowledge transferring. The simulation accuracy is significantly increased by 24.6% and the predicting error reduced by 63.6% compared with the baseline model. Moreover, compared with the other classical transfer learning modes, the method proposed has the best simulating performance on field testing data set. Furthermore, the effect study on the hyper parameters indicates that the method proposed is able to adaptively balance the weight of learning knowledge from the physical theory domain or from the real-world operation domain.     

庞庞塔煤矿综采工作面恒压供液系统研究

结合庞庞塔煤矿综采工作面的具体应用现状,设计了变频驱动控制的恒压供液系统。在泵出口管道中设置压力传感器、流量计等,同时在每台液压支架乳化液入口处安装一个蓄能器。压力、流量等数值由各种传感器来进行采集,并把监测到的数据反馈给PLC控制器。PLC控制器通过把传感器送来的数据和系统预设好的数值进行比较来判断供液系统是否需要加压或着减压,以此通过变频器来改变电机的转速,以实现供液系统维持恒压状态。     

A review on development of offshore wind energy conversion system

Wind energy conversion system, aiming to convert mechanical energy of air flow into electrical energy has been widely concerned in recent decades. According to the installation sites, the wind energy conversion system can be divided into land-based wind conversion system and offshore wind energy conversion (OWEC) system. Compared to land-based wind energy technology, although OWEC started later, it has attracted more attentions due to its significant advantages in sufficient wind energy, low wind shear, high power output and low land occupancy rate. In this paper, the principle of wind energy conversion and the development status of offshore wind power in the world are briefly introduced at first. And then, the advantages and disadvantages of several offshore wind energy device (OWED), such as horizontal axis OWED, vertical axis OWED and cross axis OWED are compared. Subsequently, several major constraints, such as complex marine environment, deep-sea power transmission and expensive cost of equipment installation faced by offshore wind conversion technology are presented and comprehensively analysed. Finally, based on the summary and analysis of some emerging technologies and the current situation of offshore wind energy utilization, the development trend of offshore wind power is envisioned. In the future, it is expected to witness multi-energy complementary, key component optimization and intelligent control strategy for smooth energy generation of offshore wind power systems.     

基于深度学习的高压油泵可靠性试验故障分析方法研究

介绍了船用柴油机高压油泵可靠性试验平台的系统架构,开展了对试验数据的清洗处理和基于机器学习的试验典型故障的研究。利用TensorFlow平台对试验中两种典型故障进行原因分析和模式识别,建立高压油泵可靠性试验故障诊断模型。结合柴油机高压油泵可靠性试验中大量反复出现的故障数据进行训练,大大提高了故障诊断模型的准确度,为高压油泵生产研制部门提供应急故障处理方案提供了依据,为实现船舶动力行业智能诊断提供了新的思路。     

某高压燃油泵滚轮浮环轴承性能分析

为了分析某高压燃油泵中的滚轮浮环轴承的性能,在考虑转轴、浮环、滚轮的弹性变形基础上,重点对比浮环轴承与常规轴承在不同工况下的承载能力、功耗、滑油流量及等效刚度阻尼等润滑特性参数的变化规律,并分析了在正常工况下浮环轴承的润滑特性,为燃油泵滚轮浮环轴承的设计提供理论基础。     

Performance evaluation of a solarized gas turbine system integrated to a high temperature electrolyzer for hydrogen production

In this paper, the performance of a solar gas turbine (SGT) system integrated to a high temperature electrolyzer (HTE) to generate hybrid electrical power and hydrogen fuel is analyzed. The idea behind this design is to mitigate the losses in the electrical power transmission and use the enthalpy of exhaust gases released from the gas turbine (GT) to make steam for the HTE. In this context, a GT system is coupled with a solar tower including heliostat solar field and central receiver to generate electrical power. To make steam for the HTE, a flameless boiler is integrated to the SGT system applying the SGT extremely high temperature exhaust gases as the oxidizer. The results indicate that by increasing the solar receiver outlet temperature from 800 K to 1300 K, the solar share increases from 22.1% to 42.38% and the overall fuel consumption of the plant reduces from 7 kg/s to 2.7 kg/s. Furthermore, flameless mode is achievable in the boiler while the turbine inlet temperature (TIT) is maintained at the temperatures higher than 1314 K. Using constant amounts of the SGT electrical power, the HTE voltage decreases by enhancing the HTE steam temperature which result in the augmentation of the overall hydrogen production. To increase the HTE steam temperature from 950 K to 1350 K, the rate of fuel consumption in the flameless boiler increases from 0.1 m/s to 0.8 m/s; however, since the HTE hydrogen production increases from 4.24 mol/s to 16 mol/s it can be interpreted that the higher steam temperatures would be affordable. The presented hybrid system in this paper can be employed to perform more thermochemical analyses to achieve insightful understanding of the hybrid electrical power-hydrogen production systems.     

Wake meandering effects on floating wind turbines

As more floating farms are being developed, the wake interaction between multiple floating wind turbines (FWTs) is becoming increasingly relevant. FWTs have long natural periods in certain degrees of freedom, and the large‐scale movement of the wake, known as wake meandering, occurs at very low frequencies. In this study, we use FAST.Farm to simulate a two‐turbine case with three different FWT concepts: a semisubmersible (semi), a spar, and a tension leg platform (TLP), separated by eight rotor diameters in the wind direction. Since wake meandering varies depending on the environmental conditions, three different wind speeds (for all three concepts) as well as two different turbulence levels (for the semi) are considered. For the below‐rated wind speed, when wake meandering was most extreme, yaw motion standard deviations for the downstream semi were approximately 40% greater in high turbulence and over 100% greater in low turbulence when compared with the upstream semi. The low yaw natural frequency (0.01 Hz) of the semi was excited by meandering, while quasi‐static responses resulted in approximately 20% increases in yaw motion standard deviations for the spar and TLP. Differences in fatigue loading between the upstream and downstream turbines for the mooring line tension and tower base fore‐aft bending moment mostly depended on the velocity deficit and were not directly affected by meandering. However, wake meandering did affect fatigue loading related to the tower top yaw moment and the blade root out‐of‐plane moment.     

Effect of rotor–tower interaction,tilt angle,and yaw misalignment on the aeroelasticity of a large horizontal axis wind turbine with composite blades

This paper presents a detailed analysis of the rotor–tower interaction and the effects of the rotor's tilt angle and yaw misalignment on a large horizontal axis wind turbine. A high‐fidelity aeroelastic model is employed, coupling computational fluid dynamics (CFD) and structural mechanics (CSM). The wind velocity stratification induced by the atmospheric boundary layer (ABL) is modeled. On the CSM side, the complex composite structure of each blade is accurately modeled using shell elements. The rotor–tower interaction is analyzed by comparing results of a rotor‐only simulation and a full‐machine simulation, observing a sudden drop in loads, deformations, and power production of each blade, when passing in front of the tower. Subsequently, a tilt angle is introduced on the rotor, and its effect on blade displacements, loads, and performance is studied, representing a novelty with respect to the available literature. The tilt angle leads to a different contribution of gravity to the blade deformations, sensibly affecting the stresses in the composite material. Lastly, a yaw misalignment is introduced with respect to the incoming wind, and the resulting changes in the blade solicitations are analyzed. In particular, a reduction of the blade axial displacement amplitude during each revolution is observed.     

轻组分绝对占优的蒸汽再压缩隔离壁蒸馏塔的最优拓扑结构

隔离壁蒸馏塔（DWC）的双塔多段拓扑结构导致蒸汽再压缩热泵（VRHP）的应用具有多种可能性，包括单VRHP、多VRHP、多级VRHP以及它们的相互组合等复杂结构，这显著加剧了蒸汽再压缩隔离壁蒸馏塔（DWC-VRHP）综合与设计的复杂性与烦琐性。为解决这一问题，针对轻组分绝对占优的三元宽沸点物系的分离问题推演了DWC-VRHP的最优拓扑结构，由此能够有效回避系统综合与设计过程中的结构搜索问题并显著降低模型化与搜索计算的工作强度。轻组分绝对占优与宽沸点物性导致了塔顶冷凝器与预分离蒸馏塔的提馏段是主要的热源与热阱，也决定了DWC-VRHP的最优拓扑结构，即一个二级VRHP与DWC的耦合系统。第一级VRHP用于进料预热，既充分利用温度提升跨度小的特点，又可以通过进料分流强化气液相间的物质传递。第二级VRHP用于加热预分离蒸馏塔的提馏段（或公共提馏段），能够最大限度地降低分离操作的非可逆性。采用苯/甲苯/邻二甲苯和正戊烷/正己烷/正庚烷两个物系的分离问题对所提出的DWC-VRHP的最优拓扑结构进行了分析与验证。通过与DWC以及DWC-VRHP其他潜在结构的系统性比较，显示了所提出系统结构在稳态性能方面的优越性。     

AISI8630钢泥浆泵排出缸失效分析及工艺改进

针对AISI8630钢锻件在实际生产中出现的锻造失效问题进行了研究。采用"EAF→LF→VD→铸锭"工艺冶炼AISI8630钢锭,钢锭锻造成泥浆泵锻件,经热处理加工后探伤发现密集性缺陷。通过化学成分、低倍分析和金相显微分析,对产生的缺陷进行分析研究。结果表明:宏观检测受检面有许多裂纹,将裂纹打开,发现断口存在明显的白点缺陷;近裂纹附近有灰色硫化物夹杂,存在明显的组织偏析现象,远离裂纹处的金相组织无明显差异。通过加强冶炼过程控制,对钢锭进行锻前消氢、消应力退火及锻后250℃以下缓冷处理等工艺优化,避免了白点缺陷的形成,并经现场验证得到合格产品。